Added awareness of the 'failed' status for an endpoint

pages/900_System_Status.py

@@ -40,6 +40,9 @@ if st_setup('LLM Arch'):
                 if st.button("Resume", key=f'action_{i}'):
                     resume_hf_endpoint('alfraser', e)
                     st.rerun()
+        elif status == HF_FAILED:
+            with status_col:
+                st.error(message)
         else:
             refresh = True
             with status_col:

src/common.py

@@ -58,6 +58,7 @@ def hf_api_token(write: bool = False) -> str:
 HF_RUNNING = 'running'
 HF_SCALEDTOZERO = 'scaleToZero'
 HF_PAUSED = 'paused'
+HF_FAILED = 'failed'
 
 
 def hf_endpoint_status(username: str, endpoint_name: str, api_token: str = None) -> Tuple[bool, bool]: